DispatchDate: 17.04.2019 · ProofNo: 133, p.1

news feature 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Credit: Dr. Robert Pope, National Biodefense Analysis & Countermeasures Center 30 31 32 33 ’s latest makeover 34 35 As issues of product consistency, standardization and specificity are being tackled, can phage therapeutics—long 36 oversold and overhyped—finally realize their antibacterial potential? Charles Schmidt investigates. 37 38 Charles Schmidt 39 40 41 n May of 2018, an international team of million grant from the UCSD chancellor, Still, previous experience, mostly in 42 researchers and clinicians reported they the new Center for Innovative Phage the context of compassionate-use phage 43 successfully treated a seriously ill teenager Applications and Therapeutics (IPATH) treatments, has shown the approach to 44 I with cystic fibrosis who had disseminated is applying “the same principles of clinical be hit-and-miss, time-consuming and 45 infection by Mycobacterium abscessus with evaluation and development to phage expensive. To turn from 46 a cocktail of genetically engineered phage1. therapy that would be applied to any other a laboratory tool into an efficacious 47 According to the University of Pittsburgh’s therapeutic entity,” says center co-director therapeutic for broader markets, companies 48 Graham Hatfull, who led the research team, Robert Schooley, a physician and infectious are seeking to scale up production and 49 this accomplishment represents a number disease specialist at UCSD. deliver potent phage products under good 50 of firsts: the first genetically engineered A worsening crisis of multi-drug- manufacturing practices (GMP) quickly 51 phage treatment—in this case, to convert resistant (MDR) infections, along with and reliably. Can companies deliver on 52 a lysogenic phage to a lytic variety—and advanced technologies for characterizing expectations? We may get the answer soon 53 the first treatment of a mycobacterium. It viruses and their host interactions, is as several companies developing phage 54 also bodes well for a therapy that has long prompting a re-evaluation of phage therapy. therapies—AmpliPhi Biosciences, Adaptive 55 been dismissed by Western practitioners, as And pharma, which has steered clear of Phage Therapeutics and Intralytix— move 56 well as for the future of synthetic-biology , let alone phage-derived ones, toward the clinic this year. 57 approaches to the vexing problem of may be taking notice. Johnson & Johnson  Q158Q2 -resistant . struck two deals centered on phage in 100 years of interlude 59 This news follows last year’s launch of a January: one with Locus Biosciences, worth First tested as a prophylactic against avian 60 phage translational research center at the upwards of $818 million, to develop CRISPR typhosis in rural France in 1919, cocktails 61 University of California, San Diego (UCSD), phages (Box 1), and the other with the Israeli of phages were used therapeutically in 62 another sign of optimism in this old but company BiomX, which is applying phage Europe and the United States during the 63 controversial approach for treating bacterial therapy to dysbiosis of the microbiome. pre-antibiotic era, and they are still prevalent 64 infections. Supported with a three-year, $1.2 in Russia and Central and Eastern Europe 65 A B C Nature | www.nature.com/naturebiotechnology DispatchDate: 17.04.2019 · ProofNo: 133, p.2

news feature 66 67 68 But funding remained scarce and validated 69 Box 1 | Turning bacterial defenses on themselves evidence of reproducible clinical efficacy 70 remained elusive, so many companies went 71 There is yet another way researchers are virulence or resistance genes. Cofounders out of business. Today, of that cadre of 72 turning phage into therapy: by loading Timothy Lu and Xavier Duportet (now companies, only GangaGen in Bangalore, 73 them up with CRISPR–Cas systems that Eligo’s CEO) showed that they could India; Baltimore-based Intralytix; and Phage 74 target host genes—that is, using the very knock out a virulent strain of S. aureus Biotech of Rehovot, Israel, are in business. 75 system that bacteria evolved to eliminate while leaving an avirulent strain intact. Intralytix and Phage Biotech survived by 76 external threats, like a phage, against itself. Likewise, they could knock out antibiotic deploying phage in agriculture to control 77 Locus Biosciences is using Cas3, resistance genes from plasmids, preventing plant diseases, detect pathogens and assess 78 which differs from the more widely used the spread of resistance10,11. They are food safety (Table 1). 79 Cas9 in that it degrades its target via positioning themselves in the microbiome 80 an exonucleolytic activity, essentially space, with their lead indication being A center of excellence 81 destroying its target gene, rather than a gut pathogenic bacteria causing a rare In 2010, Texas A&M University launched 82 introducing double-strand breaks, (undisclosed) disease. the Center of Phage Technology (CPT), 83 which prepare a template for repair or Finally, Cambridge, UK, start-up with Ry Young, professor of biochemistry, 84 replacement. Cofounder and researcher at Nemesis Biosciences is developing a biophysics and biology, at the helm. Like 85 North Carolina State University Rodolphe platform of so-called ‘transmids’ that many of the companies, Young says CPT’s 86 Barrangou and colleagues published a borrows elements of phage biology to initial strategy was to focus on agriculture 87 foundational paper in 2013 in which they deliver RNA-guided nucleases to knock out and animal husbandry, “while avoiding 88 show that Cas systems targeting specific antimicrobial resistance genes. Transmids human involvement because of high 89 bacterial sequences can distinguish among carry only a short signal phage sequence regulatory loads.” Now that’s changing. 90 closely related species and can control the that is needed for packaging into a phage Like other facilities that work in this area, 91 number of individual strains in a mixture9. capsid. They are grown in a helper bacterial the CPT is overwhelmed with requests for 92 According to senior VP Joseph Nixon, host that harbors a defective phage particle. phages as treatment of last resort for patients 93 CRISPR–Cas3 efficiently kills target cells After being purified from the host bacteria, with MDR infections. Indeed, the recent 94 regardless of the function of the targeted the transmids invade the pathogenic resurgence in human interventions and 95 sequence, and multiplexing is possible, bacteria in what CSO Conrad Lichtenstein emergency investigational new drug (eIND) 96 which, like phage cocktails, forestalls the calls a “hit and run” attack: transmid DNA applications “has come out of nowhere as a 97 development of resistance. Their first trial, enters the host cell, replicates and expresses spontaneous reaction to the MDR problem,” 98 which is in the planning stages, could be the nuclease, inactivating resistance genes. Young says. 99 among the first controlled, randomized In this way, the strategy avoids waves of The CPT supplied phages for a highly 100 trials of bacteriophage therapy and the first phage infection against which the patient publicized eIND treatment in 2016. Tom 101 using engineered phage. might mount an immune response, and, Patterson, a UCSD professor of psychiatry, 102 Eligo Bioscience in Paris is developing since it avoids killing directly, the bacterial had picked up an MDR Acinetobacter 103 a non-replicative phage platform that pathogens are not under direct selection to baumannii infection during a trip to Egypt 104 carries Cas9 targeted to bacterial evolve resistance, Lichtenstein says. and was near death from organ failure 105 by the time his therapy was delivered. 106 Patterson’s wife, Steffanie Strathdee, an 107 epidemiologist at UCSD, helped coordinate 108 the phage treatments that ultimately saved 109 today, for wound infections, gastroenteritis, Then in the early 2000s the field her husband’s life3. Contributed by three 110 sepsis and other ailments. In the West, sputtered back into life. The rise of modern separate entities—the CPT; AmpliPhi 111 phage therapy was abandoned after broad- sequencing technology began to revive Biosciences; and the US Naval Medical 112 spectrum antibiotics came on the scene. interest in phage biology and suggested that Research Center (NMRC), which has been 113 Later, during the 1940s, the California molecular engineering and characterization researching phage since 2011 for possible 114 Institute of Technology’s Seymour Benzer could start to address such issues as battlefield use—the phages were given 115 pioneered work with bacteriophage T4 as product consistency (phage cocktails often daily for two months through a catheter 116 an experimental genetic tool, leading to contain more than ten phage strains), poor into Patterson’s abdomen, as well as 117 the deciphering of the central dogma of tissue distribution, pharmacodynamics through an intravenous line. Within days 118 molecular biology and spurring its rise as a and immunogenicity issues. Despite the of beginning treatment, Patterson awoke 119 field. However, for a variety of reasons—the difficulty of gaining intellectual property from a coma, and after three months he 120 preponderance of phage work on laboratory protection (the technology has been around was cleared of infection. He remains in 121 rather than pathogenic bacteria (which for nearly a century), the skepticism of good health. Meanwhile, Strathdee, who 122 often are already naturally resistant to traditional pharma and venture investors, is the associate dean of global health 123 ) and a lack of peer-reviewed and the uncertainties surrounding clinical science at UCSD, and Schooley, who was 124 published data showing efficacy in animal trials and regulatory oversight, a clutch of Patterson’s treating physician, joined forces 125 models, not to mention political bias companies sprang up around the concept by to launch IPATH. 126 against a science associated with the Soviet the mid-2000s, including Biophage Pharma, Aside from one-off treatments like 127 Union and Nazi Germany—meant that the Enzobiotics, Exponential Biotherapies, Patterson’s, the hope is that phages will 128 therapeutic utility of these viruses remained GangaGen, Hexal Gentech, Intralytix, find broad use as precision antimicrobials 129 largely unexplored; research funding in MicroStealth Technologies, Phage Biotech, that target specific pathogens while 130 the field dropped off during the 1970s and Phage Therapeutic, PhageGen, PhageTech, sparing the healthy members of the human 131 remained flat for decades. Phage Therapy and Phico Therapeutics2. microbiome. The treatments can also be A B C Nature Biotechnology | www.nature.com/naturebiotechnology DispatchDate: 17.04.2019 · ProofNo: 133, p.3

news feature 132 133 134 Table 1 | Selected companies developing phage therapies 135 136 Company Founded Funding Platform Target Status 137 Adaptive Phage 2016 $5 million seed Individualized phage Infectious disease, Preclinical 138 Therapeutics therapy, PhageBank, Host urinary tract infections 139 Range Quick Test 140 AmpliPhi Biosciences 2002 $8.23 million Phage combinations for Infectious disease Phase 1 cocktail of three phages, 141 market bacteria skin safety test; individual access 142 capitalization for S. aureus and P. aeruginosa 143 BiomX (Ness Ziona, Israel) 2015 $24 million Customized phage Irritable bowel disease Preclinical 144 cocktails 145 C3J Therapeutics 2005 $136 million Antimicrobial peptides Infectious disease, Preclinical 146 and engineered phages microbiome 147 148 Eligo Bioscience (Paris) 2014 $20.2 million CRISPR engineered phage infectious disease Preclinical 149 series A (Khosla, 150 Seventure) 151 EnBiotix 2012 Not disclosed Engineered phage Joint, skin, wound, Preclinical 152 cystic fibrosis, prosthetic joint 153  Q3 infections 154  155 Intralytix 1998 $17.5 million Phage cocktail against Crohn’s disease Phase 1/2 156 adherent, invasive E. coli 157 Locus Biosciences 2015 $26 million CRISPR engineered phage Infectious disease, Preclinical 158 microbiome 159 Nemesis Biosciences 2014 $2.3 million Transmid Extended spectrum Preclinical 160 (Cambridge, UK) β-lactamase- 161 producing bacteria 162 Pherecydes Pharma 2006 $12.3 million Individualized phage Infectious disease Phase 2 (burns) 163 (Romainville, France) therapy 164 165 166 167 168 used in combination with antibiotics. Phage hunts lab ten years ago4 to extend the host range 169 Indeed, evidence shows that surviving Phage therapy has been benefitting from of one isolate and convert another from 170 bacteria following phage treatment are in technical advances on multiple fronts. As lysogenic to lytic. 171 some cases resensitized to antibiotics and Rotem Sorek, of the Weizmann Institute Recently, researchers at the University of 172 incur additional fitness costs that render of Science in Rehovot, Israel and one of California, Berkeley have received internal 173 them susceptible to immune cells. MDR the founders of BiomX, puts it, where funding from the university’s Innovative 174 A. baumannii cells, for instance, are researchers once lacked the means to Genomics Institute to create aphage foundry. 175 shielded in a capsule that protects them characterize phages rapidly, it’s now possible Here the goal is to identify genes involved 176 from antibiotics. The surviving bacteria to “sequence a phage, assemble its genome in phage–host interactions, using loss-of- 177 during Patterson’s treatment lacked this and analyze the results in a matter of days.” function mutational analysis5 and gain-of- 178 capsule and succumbed to antibiotic Next-generation sequencing has expanded function gene dosage effects to identify the 179 treatment. the numbers of viral sequences deposited in bacterial genes that enable particular phages 180 But despite the fact that phage have been databases. Since the 1990s, the University to bind or resist binding by others. These 181 a subject of research for nearly a century, of Pittsburgh’s Hatfull has been collecting insights would enable engineering of phages 182 very little is known about them, outside of actinobacteriophages as part of his long- and phage-like systems to target specific 183 the few research workhorse species. Even for standing interest in mycobacteria (a genus groups of microbes and deliver designed 184 these comparatively simple organisms, for of Actinobacteria). It was slow going at function, and conversely, engineering of 185 many phage species the function of as much first—the first genome sequence took a possible hosts for particular resistance and 186 as 90% of the genome remains unknown year. But the pace has quickened, and the sensitivity phenotypes. According to joint 187 function. “So you can get a readout of the database, PhagesDB, which is nourished by principal investigator Vivek Mutalik, the 188 genes, but unfortunately most of the genes an annual phage hunt, has collected over focus initially will be on eliminating plant 189 are annotated as hypothetical. No idea 15,000 isolates, of which over nearly 3,000 and human pathogens, and later efforts will 190 what they are doing,” says Karen Maxwell, have been sequenced completely. In fact, it involve rationally engineering microbial 191 a biochemist and phage specialist at the was from this cache of phage that the three- communities involved in plant health and 192 University of Toronto. Maxwell cautions that phage cocktail used to treat the patient with productivity, those that remediate heavy- 193 phage, which are notoriously promiscuous cystic fibrosis was derived. Hatfull says metal contaminants at complex watersheds, 194 and can swap genes among phage species finding and preparing the phage was not and those populating the human gut. 195 as well as hosts, could be carrying virulence trivial: after finding isolates that infected Identifying phage target receptors is key 196 genes or genes for bacterial toxins or M. abscessus, they used a phage-based to moving phage into the clinic, as well as 197 antibiotic resistance. recombineering technique developed in his to preventing the host from developing A B C Nature Biotechnology | www.nature.com/naturebiotechnology DispatchDate: 17.04.2019 · ProofNo: 133, p.4

news feature 198 199 200 201 Meeting the clinical benchmark 202 Accompanying these technical advances are 203 encouraging signals from regulators at the 204 US Food and Drug Administration (FDA). 205 During an FDA-sponsored workshop on 206 phage therapy held in 2017, Scott Stibitz, a 207 lab chief in the agency’s Center for Biologics 208 Attachment Evaluation and Research, said that after 209 Lysis many years of largely anecdotal experience, 210 the field must now “initiate scientifically 211 Induction rigorous programs that include adequate and 212 well-controlled clinical trials that support 213 licensure of phage therapy products,” a 214 Cellular Lysogenic replication process that, he said, “FDA is committed 215 cycle to facilitating.” The principal concerns for 216 Assembly Prophage DNA injection FDA for phage preparations are that they 217 Prophage are safe, pure, potent, non-lysogenic, non- 218 establishment transducing, free of undesirable genes, and 219 low in bacterial endotoxins, which can 220 contaminate phage lysates extracted from 221 bacterial hosts. 222 Lytic cycle As Stibitz made those comments, the first 223 clinical trial with GMP-compliant phage 224 Replication therapy was already underway. Launched 225 in 2013 by Pherecydes Pharma, a biotech 226 company in Romainville, France, the 227 Transcription multicenter PhagoBurn trial tested a cocktail 228 and translation of 12 phages in burn patients who had 229 developed MDR P. aeruginosa infections. 230 Phage can infect host bacteria in two distinct pathways: lytic cycle, in which new phage are created Unfortunately, the results, which were 231 and dispersed, and lysogenic, whereby the phage and host form a stable association. Reprinted with published last October, were both lackluster 8 232 permission from ref. , Springer Nature. and disappointing for phage proponents7. 233 The patients had been randomized to 234 either phage therapy or a control antibiotic, 235 resistance, notes Young. By combining burden and potentially resensitize bacteria sulfadiazine silver, both given in topical 236 phages into cocktails, each targeting a to chemical antibiotics and reduce the emulsions. Although the phages showed 237 different viral receptor, it’s possible to expression of extracellular virulence factors,” activity—bacterial loads fell substantially in 238 prevent the onset of resistance, he says. says Chan. the treated patients—reduction were not as 239 Benjamin Chan, at the Department of At the same time, Yale’s phage library, fast as in the control group, so the trial was 240 Ecology & Evolutionary Biology at Yale in addition to those at other institutions, terminated prematurely. Jérôme Gabard, 241 University, adds that deep sequencing can is amassing collections of characterized, the chief operating officer at Pherecydes, 242 allow clinicians to monitor in real time purified phages for clinical use—each says the treatment wasn’t as effective as 243 whether treatment is reducing the target of them targeted at known bacterial hoped because the number of phages in 244 bacterial population, and annotation tools receptors—which can then be pulled off the the cocktail was too high. The aim had 245 can identify sequences for undesirable shelf and made available for treatment6. been to create an off-the-shelf cocktail with 246 genes, such as integrases, that enable phage That capacity marks progress over the enough different phages to fully address P. 247 to integrate within bacterial genomes and ad hoc experience of treating Patterson’s aeruginosa’s intra-species diversity and its 248 remain dormant. Chan curates a growing infection, during which “people worked assortment of receptors. Each of 12 phages 249 phage library housed in the laboratory of round the clock for three weeks to was individually stable, but when mixed 250 Yale professor Paul Turner. By using these identify phages that matched his isolate and together, they reacted in ways that Gabard 251 tools to assess patient samples during the then grow and purify them,” according to says aren’t well understood. 252 course of therapy, “we can see changes in Strathdee. And that wasn’t the end The phages were combined months 253 bacterial populations at the genomic level, of it. Patterson developed resistance to his before the patients were treated, during 254 check for bacterial resistance patterns, and initial treatment after two weeks, which time phage titers in the mixture fell 255 determine how phages adapt to them,” he forcing researchers to hunt for new by up to five orders of magnitude, resulting 256 says. With over 500 phages for over a dozen therapeutic candidates in sewage samples, in delivered doses that were far lower than 257 human pathogens, the library has already which then had to be characterized intended. The titers were reduced further 258 been mined for phage to be applied in and amplified in sufficient quantities still after Pherecydes diluted its cocktail with 259 upcoming clinical trials. They are planning for dosing. The saga of Patterson and saline to address a growing problem with 260 a clinical trial at Yale New Haven Hospital Strathdee has been documented in a endotoxin contamination. 261 to treat MDR Pseudomonas aeruginosa forthcoming book, The Perfect Predator According to Young, the PhagoBurn 262 infections associated with cystic fibrosis. (https://bioengineeringcommunity.nature. team made another critical error: they never 263 “Our objectives are to reduce bacterial com/). identified the target receptors for each phage A B C Nature Biotechnology | www.nature.com/naturebiotechnology DispatchDate: 17.04.2019 · ProofNo: 133, p.5

news feature 264 265 266 approach. “It’s about phages and antibiotics.” 267 Should FDA approval follow, the company 268 intends to market an off-the-shelf product 269 prepared at its GMP-certified facility in 270 Slovenia and then monitor for resistance 271 with post-market surveillance that screens 272 the phages against isolates sampled from 273 different countries over time. Grint expects 274 that, should resistance be detected, they 275 will address it by tweaking the mixture 276 and periodically bringing revised products 277 to market, not unlike the way new flu 278 vaccine are made available every year. The 279 company’s three-phage cocktail targeting 280 S. aureus has undergone two phase 1 trials 281 for topical application, and this cocktail, 282 as well as one for P. aeruginosa, is available 283 for expanded access for patients with life- 284 threatening infections. In October, they 285 reported clinical case series data obtained 286 from 13 Australian patients with MDR S. 287 aureus sepsis and/or bacteremia. Ten of the 288 patients also had infective endocarditis, and 289 in half of them, the infection has moved 290 Phage researchers Mei Liu, Ry Young, Carlos Gonzalez and Jason Gill from the Center for Phage into prosthetic valves. Collectively, over 290 291 Technology at Texas A&M take the measure of the phage. Since 2010, the center has been working on intravenous phage doses were administered,  292 Q4 and 83% of the intent-to-treat population phage therapeutics for humans, animals and plants. Credit: Mark Guerrero, Division of Marketing & 293 Communications, Texas A&M University (10 of 12 patients) were either symptom-free 294 or significantly improved by the end of the 295 treatment. The company’s planned trials for 296 before the viruses were combined. Adding pathogens are more genetically diverse. later this year will test cocktails targeting 297 two phages targeted against the same As occurred during Patterson’s treatment, two different indications: MDR S. aureus 298 receptor wouldn’t make sense, he says, “since infectious isolates in these cases have to be infections in implanted vascular assist 299 resistance to one automatically confers monitored continually for resistance. And devices and P. aeruginosa associated with 300 resistance to the other.” The appropriate when a given phage, either in isolation or cystic fibrosis. 301 number and combination of phages varies combined with other phages in a cocktail, An upcoming clinical trial from Adaptive 302 with the genetic diversity of the target stops working, a new one targeted at a Phage Therapeutics (APT), meanwhile, 303 pathogen. Bacterial species with low genetic different receptor can be added in its place. will test a more personalized strategy for 304 diversity and a correspondingly limited suite Clinical trials planned by two companies urinary tract infections. The company’s 305 of phage receptors—Staphylococcus aureus, this year illustrate the two approaches. In cofounder and chief science officer, Carl 306 for instance—can be treated with just a few collaboration with IPATH, AmpliPhi has Merril, spent over four decades trying to 307 phages. As the genetic diversity of the target plans to test a fixed cocktail of three lytic bring phage therapy forward while working 308 pathogen increases, however, so too does phages targeted against S. aureus infections as a lab chief at the US National Institutes of 309 the diversity of phages needed for effective located within ventricular assist devices, Health. Rebuffed with repeated cuts to his 310 therapy. P. aeruginosa is intermediate in infections that are hard to treat. Paul funding, Merril retired in frustration from 311 diversity, whereas A. baumannii varies so Grint, the company’s CEO, says the fixed the NIH in 2005, while his postdoc, Biswajit 312 widely from strain to strain that dozens cocktail is more amenable to a conventional Biswas, persevered with phage research, 313 of different phages are needed to treat it development program, and it’s also “more first in the private sector and then at the 314 successfully. consistent with our goals for a product NMRC, at Fort Dietrick, , where 315 that’s stable in a refrigerator and addresses he is now chief of bacteriophage sciences. 316 Differing testing strategies the many issues FDA has with quality and Since arriving at the NMRC, Biswas has 317 The link between bacterial diversity and reproducibility.” The company tested over been building up a library of phages 318 phage combinations is fundamental to 100 phages before selecting three candidates collected from around the world, now called 319 the strategies that companies are exploring with a 95% kill rate against a panel of S. PhageBank, which was inspired initially by 320 as they enter the clinic. One strategy aureus strains. According to Grint, the the need for better therapy against MDR A. 321 employs fixed cocktails with a minimal phages induce the production of enzymes baumannii infections that occur commonly 322 number of phages (Young recommends that disrupt bacterial biofilms, which tend among injured military personnel returning 323 no more than three or four) targeting to grow on prosthetic implants and shield from the Middle East. To speed the phage 324 low-diversity bacteria. The phages in this pathogens from antibiotics. Grint says screening process, he developed a high- 325 case can be pulled out of a refrigerator and phages could therefore expose bacteria to throughput 96-well system, the Host 326 used for treatment, much like any other antibiotic treatment, and thereby potentiate Range Quick Test (HRQT), which uses a 327 therapeutic agent. their action. colorimetric assay to assess the effects of 328 Another strategy is personalized and “It’s not about phages or antibiotics,” phage on bacterial growth and survival. 329 geared for patients whose infectious Grint says of the company’s therapeutic Strathdee had enlisted NRMC’s help during A B C Nature Biotechnology | www.nature.com/naturebiotechnology DispatchDate: 17.04.2019 · ProofNo: 133, p.6

news feature 330 331 332 lead product, sponsored initially by a grant 333 from Mayo Clinic Ventures, is engineered 334 to accelerate biofilm degradation7 while at 335 the same time targeting S. aureus–induced 336 prosthetic joint infections. Jeff Wager, the 337 company’s CEO, says this fixed product is 338 about two years away from the clinic. 339 Finally, phage are also being coopted 340 as delivery vehicles for other therapeutic 341 modalities. This has spurred a cadre of 342 companies exploiting molecular tricks to turn 343 bacteria’s own CRISPR–Cas endonuclease 344 immune systems on themselves (Box 1). 345 346 Au naturel or modified? 347 Engineered phage may have certain 348 advantages over natural ones, particularly 349 from a standpoint of commercial 350 development. Collins points out that

351  “engineered modifications” are “eminently Q5 Adaptive Phage Therapeutics was founded to move technology developed at the US Navy’s Biological 352  patentable and natural phages likely not.” 353 Defense Research Directorate phage research from compassionate use into commercially available Even so, Biswas questions the wisdom 354 therapy. Credit: Carl Merril, Adaptive Phage Therapeutics of spending years on patenting genetic 355 modifications to which bacteria might easily 356 develop resistance. And Greg Merril points 357 her husband’s ordeal in 2016. Galvanized by major hospitals, each stocked with single- out that while they’re unable to patent viral 358 his successful recovery, the NMRC began use phage vials manufactured under GMP. sequences, companies working with natural 359 looking into commercial opportunities with phages can pursue other forms of intellectual 360 the PhageBank and HRQT, and APT was Phage engineers property. APT, for instance, has a portfolio 361 subsequently formed by Merril and his son AmpliPhi and APT are working with natural of pending patents related to banking phage, 362 Gregory Merril (now the company’s CEO) phages, but several other companies have high-throughput bacterial-phage matching, 363 for that purpose. recently sprung up that are engineering and in silico phage matching with artificial 364 APT entered into a multi-year the viruses to deliver enhanced therapeutic intelligence. Similarly, AmpliPhi has patents 365 cooperative research and development payloads. C3J Therapeutics (which in on “properties, selection, and combination 366 agreement with the NMRC in 2016 and January announced plans to merge with of natural phages,” Grint says. 367 was awarded exclusive license to the HRQT AmpliPhi) acquired the proprietary phage But even as companies move toward 368 and the PhageBank the following year. The platform of Synthetic Genomics in February clinical trials, they’re confronting entrenched 369 PhageBank is split between the company’s 2018, along with what Brian Varnum, biases against phage therapy by physicians 370 facility in Gaithersburg and the NMRC. C3J’s chief development officer, says is inclined to view it as an old Soviet technology 371 Infectious isolates delivered to APT by the a lead program targeting P. aeruginosa that was never backed by reliable evidence. 372 treating hospital can be screened against that he anticipates will reach the clinic in Phages may offer the promise of selective 373 the contents of the PhageBank to search for 2019. C3J’s strategy is to hunt for natural anti-infective activity, but using them requires 374 effective candidates. phages with broad host ranges and then that physicians first identify the species, 375 “Say the patient has a Klebsiella infection,” engineer them for desired attributes, such which in a typical hospital setting might take 376 Biswas says. “There are hundreds of phages as improved pharmacology, greater depth 24–48 h, if not more. “What you really need 377 targeting that pathogen in the PhageBank, of kill and greater access to biofilms. The to be successful in this space is phage plus 378 and you can screen them all.” As resistance company plans to develop fixed cocktails. a rapid diagnostic kit,” says Aleks Radovic- 379 emerges during treatment, phages can Though Varnum acknowledges that fixed Moreno, vice president of PureTech Health. 380 simply be swapped out, explains Mike cocktails likely won’t work against highly Though sequencing technologies can identify 381 Stockelman, deputy director for infectious diverse pathogens such as A. baumannii, bacteria to species within hours, they haven’t 382 diseases at the NMRC. “The whole premise he proposes that with synthetic biology been widely adopted in clinical settings due 383 behind this approach is that a cocktail works it’s possible to engineer phages that can to their high cost. “Most of the time, you can 384 for as long as it does, and when it stops, you overcome host-range limitations. give antibiotics and there isn’t a problem,” 385 go back to the assay and run it again—we Also working on engineered phages, as Radovic-Moreno says. “So the urgency isn’t 386 can do that in eight hours—and develop a one of a number of strategies for combatting there—you ask doctors and they’ll tell you 387 second cocktail. And you continue doing antibiotic resistance, is EnBiotix, in Boston, ‘Yes, MDR bacteria are an issue we need to 388 for as long as it takes to knock down an deploying technology developed by James deal with,’ but ask them to buy a $500,000 389 infection to the degree that the patient’s Collins, who is now at the Massachusetts box and take a chance on phage for one out 390 immune system can take over.” Institute of Technology. The company of every 100 patients, and it’s going to be 391 As the NMRC’s commercial partner, acquires phages from its global network of difficult.” Until there’s a pull from the clinical 392 APT is negotiating the regulatory processes academic laboratories and engineers them side, Radovic-Moreno predicts “phage 393 involved in turning the PhageBank into for a range different payloads, including companies [that don’t provide diagnostics] 394 an FDA-approved product. Merril says the antimicrobial peptides, antitoxins, and genes are going to be pushing the boulder up a very 395 long-range goal is to set up APT kiosks in for reversing phage resistance. EnBiotix’s steep hill for a long time.” A B C Nature Biotechnology | www.nature.com/naturebiotechnology DispatchDate: 17.04.2019 · ProofNo: 133, p.7

news feature 396 397 398 Yet Schooley strikes a more optimistic than it ever was before: screening is faster Published: xx xx xxxx 399 tone, and says that he and others at IPATH and cheaper, he says, and scientists have https://doi.org/10.1038/s41587-019-0133-z 400 view themselves as “filling a black hole that’s made key strides in their understanding of 401 been plaguing phage therapy for years.” phage pharmacology. At the same time, GMP References 402 IPATH has a full plate, he says, working production costs pose a major hurdle, and 1. Dedrick, R. M. et al. Nat. Med. https://doi.org/10.1038/s41591- 019-0437-z (2019). 403 with the Antibacterial Resistance Leadership Young says scientists still have a long way to go 2. Thiel, K. Nat. Biotechnol. 22, 31–36 (2004). 404 Group of the US National Institute of Allergy in terms of characterizing basic phage biology 3. Schooley, R. T. et al. Antimicrob. Agents Chemother. 61, e00954–17 405 of Infectious Diseases and with several biotech and its interactions with pathogenic bacteria. (2017). 406 4. van Kessel, J. C. & Hatfull, G. F. Nat. Methods 4, companies on what’s shaping up to be the first “Phages are viruses and they will do what 147–152 (2007). 407 wave of randomized clinical and translational they want to do and not necessarily what we 5. Price, M. N. et al. Nature 557, 503–509 (2018). 408 studies focused on phage therapeutics in the want them to,” he says. “So ultimately when 6. Chan, B. K. et al. Evol. Med. Public Health 2018, 60–66 (2018). 409 7. Jault, P. et al. Lancet Infect. Dis. 19, 35–45 (2019). United States. The first of these studies—its phage are widely used—and I think they will 8. Salmond, G. P. & Fineran, P. C. Nat. Rev. Microbiol. 13, 777–786 410 collaboration with AmpliPhi targeting S. be—it has to be on firm scientific footing.” ❐ (2015). 411 aureus–infected vascular assist devices—was 9. Gomaa, A. A. MBio https://doi.org/10.1128/mBio.00928-13 412 announced in January. Charles Schmidt (2014). 10. Citorik, R. J., Mimee, M. & Lu, T. K. Nat. Biotechnol. 32, 413 Schooley says phage therapy is better Portland, Maine, USA. 1141–1145 (2014). 414 positioned now to make broad advances e-mail: [email protected] 11. Citorik, R. J. et al. Nat. Biotechnol. 32, 1141–1145 (2014). 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 A B C Nature Biotechnology | www.nature.com/naturebiotechnology QUERY FORM

Nature Biotechnology Manuscript ID [Art. Id: 133] Author Charles Schmidt

AUTHOR:

The following queries have arisen during the editing of your manuscript. Please answer by making the requisite corrections directly in the e.proofing tool rather than marking them up on the PDF. This will ensure that your corrections are incorporated accurately and that your paper is published as quickly as possible.

Query No. Nature of Query

Q1: Author surnames have been highlighted - please check these carefully and indicate if the first name or surname have been marked up incorrectly. Please note that this will affect indexing of your article, such as in PubMed.

Q2: Please note that the eproof should be amended in only one browser window at any one time, otherwise changes will be overwritten.

Q3: Please confirm EnBiotix indications. I can’t tell from the comment thread how it should read.

Q4: “the center has been working on phage therapeutics” ok as edited?

Q5: “Adaptive Phage Technologies” changed to “Adaptive Phage Therapeutics”; correct?

Springer Nature